REVIEW 2 major objections 6 minor 39 references
First search for solar axions produced by resonant transverse-plasmon conversion finds no signal and places the first experimental bound on this production channel.
Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →
T0 review · deepseek-v4-flash
2026-08-01 08:59 UTC pith:TVKPYAET
load-bearing objection First experimental search for transverse-plasmon solar axions; null result is clean, but the headline limit rests on a solar B-field model that saturates upper bounds and the printed modulation equations have a factor-2ε typo. the 2 major comments →
First Experimental Bounds on Transverse Plasmon Solar Axions with ANAIS-112
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
For each axion mass, transverse-plasmon–axion conversion happens on a resonant shell where the local plasma frequency equals the axion mass, so the flux and its spectrum are set by the magnetic field and temperature at that specific solar radius. No statistically significant annual modulation is observed in the 1–20 keV window; the null result is converted into an exclusion curve. The key number is g_aγ < 1.32×10⁻⁹ GeV⁻¹ (90% C.L.) at m_a ≈ 150 eV, a factor of about 2.73 improvement over using the Primakoff flux alone at that mass. The exclusion band in the 35–280 eV range owes its shape to the resonance condition, and the experiment's sensitivity there is dominated by the transverse-plasmon
What carries the argument
The central mechanism is resonant transverse-plasmon–axion conversion: inside the Sun, photons acquire an effective mass equal to the local plasma frequency ω_p(r), and axions of mass m_a convert most efficiently in a thin shell where ω_p(r) = m_a. The conversion probability is a Breit–Wigner resonance regularized by the transverse photon damping rate, and the flux scales as g_aγ² times the square of the local magnetic field with a 1/3 angular average. Detection uses the inverse Primakoff effect in NaI(Tl) crystals, and the signal is extracted from the annual modulation arising from the 1.67% eccentricity of Earth's orbit, with the period fixed to one year and the phase fixed to perihelion.
Load-bearing premise
The predicted transverse-plasmon axion flux assumes a solar radiative-zone magnetic field peaking around 3×10⁷ G, but the real field is unmeasured and the paper itself notes that the adopted model saturates observational upper limits rather than being a measurement—if the field is weaker, the flux and the quoted coupling bounds scale down.
What would settle it
Measure the internal solar magnetic field in the resonance region, for example through improved helioseismic inversions or solar oblateness data, and recompute the transverse-plasmon flux; alternatively, run a helioscope with sensitivity around g_aγ = 1.3×10⁻⁹ GeV⁻¹ at m_a ≈ 150 eV—a detected line near 150 eV would confirm the mechanism, while a null result together with an independently measured field near 3×10⁷ G would refute the current interpretation.
If this is right
- A previously unprobed solar axion source is now experimentally constrained, with the strongest new limit g_aγ < 1.32×10⁻⁹ GeV⁻¹ at 90% C.L. around m_a ≈ 150 eV.
- In the 35–280 eV mass range, the transverse-plasmon flux dominates the Primakoff flux by up to two orders of magnitude, improving the excluded coupling by up to a factor of 2.73.
- All fitted modulation amplitudes between 1 and 20 keV are consistent with the null hypothesis, so no positive inverse-Primakoff solar axion signal appears in this 625.75 kg·yr exposure.
- The shape of the exclusion curve maps the solar interior: each axion mass probes the magnetic field and plasma at a different resonance radius, so the limit curve encodes information about the radiative-zone field.
- Existing low-background underground experiments can already test this mechanism without building new detectors.
- The transverse-plasmon signal peaks near 2 keV, whereas the Primakoff signal peaks near 6 keV, so spectral information can help separate the two production channels in future analyses.
Where Pith is reading between the lines
- Because the transverse-plasmon flux scales as the square of the magnetic field, the same null data could be recast as an upper limit on the radiative-zone magnetic field for a fixed coupling, offering a way to probe the solar interior that is independent of helioseismology.
- The claim's reach depends on the solar magnetic-field model; if the true radiative-zone field is much weaker than the assumed peak near 3×10⁷ G, the quoted limits would weaken proportionally, so an independent measurement of that field would sharpen or challenge the result.
- A helioscope with sensitivity below g_aγ ≈ 1.3×10⁻⁹ GeV⁻¹ in the 35–280 eV mass range could directly confirm resonant transverse-plasmon production, turning the current exclusion into a detection or ruling out the assumed flux normalization.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports a search for solar axion-like particles produced via resonant conversion of transverse plasmons in the Sun's magnetic field, using 625.75 kg·yr of ANAIS-112 data. The search exploits the annual modulation of the solar axion flux caused by the Earth–Sun distance variation, with detection through the inverse Primakoff effect in NaI(Tl) crystals. No statistically significant modulation is found in 1–20 keV bins, and the authors derive the first experimental constraints on the axion–photon coupling that include the transverse-plasmon production channel, with the strongest limit g_aγ < 1.32×10^-9 GeV^-1 (90% C.L.) at m_a ≈ 150 eV. The signal prediction combines the conventional Primakoff flux with the resonant transverse-plasmon flux using the SolarAxionFlux package and a solar magnetic-field model chosen to saturate helioseismological upper bounds.
Significance. If the result stands, it opens a genuinely new experimental window: solar axion production through resonant transverse-plasmon conversion has not been searched for before, and the ANAIS-112 exposure already probes this channel. The null modulation measurement itself appears robust: the analysis uses a detailed time-dependent background model, simultaneous fits over nine detectors, and a large exposure. A further strength is the use of the open-source SolarAxionFlux package and the availability of the analysis code, which makes the flux calculation reproducible. However, the central limit is conditional on an unmeasured solar magnetic-field normalization, and the printed signal-modulation equations are internally inconsistent. These issues must be resolved before the claimed 'first experimental bounds' can be accepted as stated.
major comments (2)
- [Eqs. (8)–(9)] These two equations are inconsistent. Expanding the Earth–Sun distance to second order gives R(t) = R_a [1 + 2ε cos(ω(t−t0)) + 3ε^2 cos^2(ω(t−t0)) + ...]. If Eq. (8) is written as S(t) = A [2ε cos(...) + 3ε^2 cos^2(...)], then the parameter A must be R_a. Equation (9), however, defines A ≈ (R_a/2)(4ε) = 2ε R_a to leading order. Substituting Eq. (9) into Eq. (8) suppresses the first-harmonic coefficient by a factor 2ε ≈ 0.033 relative to the physical flux modulation. Since the expected signal scales as g_aγ^4, this would shift the derived coupling limit by roughly (2ε)^−1/4 ≈ 2.3 if Eq. (9) were actually used. The authors should correct Eq. (8) (or Eq. (9)) and state explicitly which convention was used in the numerical pipeline; otherwise the quoted limit is affected at a level far exceeding the stated systematics.
- [Supplemental Material, 'Solar Magnetic-Field Models'; Eq. (3)] The normalization of the transverse-plasmon flux is proportional to |B(r_res)|^2, and the headline limit g_aγ < 1.32×10^-9 GeV^-1 at m_a ≈ 150 eV is set by the resonant flux from the radiative zone. The Supplemental Material states that the adopted 'seismic' model is 'chosen to saturate these observational limits and should not be interpreted as measurements of the internal solar magnetic field.' Helioseismology and oblateness provide upper bounds only, so the true radiative-zone field could be substantially lower. If the peak field were 3×10^6 G instead of 3×10^7 G, the TP flux drops by two orders of magnitude and the coupling limit weakens by a factor ~√10 ≈ 3.2, which would erase most of the claimed TP enhancement over the Primakoff-only limit. The quoted +19%/−10% systematic from the magnetic-field model samples only models that saturate the observational upper bounds and does not in
minor comments (6)
- [Supplemental Material, Table I] Table I reports the dominant systematic uncertainty 'for m_a = 0', but the headline limit is at m_a ≈ 150 eV. Since different axion masses resonate at different solar radii and therefore in different magnetic-field regimes, the systematic should be evaluated at the mass of interest.
- [Fig. 3] The label 'null χ2 = 1.32' should specify whether this is a reduced chi-square and give the number of degrees of freedom, so the reader can assess the goodness of fit.
- [Ref. [11]] In reference [11], the author name 'M. Gianotti' appears to be a typo for 'M. Giannotti' (as spelled in the author list of this manuscript).
- [Supplemental Material, Fig. D caption] The phrase 'TPC exclusion limit' in the Figure D caption should probably read 'TP' (transverse-plasmon) exclusion limit, or otherwise the acronym should be defined.
- [Eq. (7)] Equation (7) mixes differential flux, cross-section, numbers of nuclei, target mass, exposure, and detector response in a way that is dimensionally unclear. Please clarify whether this is a rate or an expected count, and define F_det precisely.
- [Bayesian analysis paragraph after Fig. 3] The prior on g_aγ^4 is described as 'flat'. Since the signal strength is proportional to g_aγ^4, this is a specific prior choice and not equivalent to a flat prior on g_aγ itself. A short justification, or a statement of how the upper limit depends on this choice, would be helpful.
Circularity Check
No significant circularity: the signal prediction is external and open-source, and the measured modulation amplitudes are not recycled into the prediction.
full rationale
The paper's derivation chain is not circular. The predicted solar axion signal is built from two external inputs: the Primakoff flux computed with the open-source SolarAxionFlux package [19] and the transverse-plasmon flux given by Eq. (3), taken from the published calculation of Ref. [11] and implemented in the same package. Neither input is fitted to the ANAIS-112 modulation data. The measured annual modulation amplitudes are extracted independently from the background-plus-modulation fit, and the resulting null result is then compared with the predicted amplitudes to set limits on g_aγ. No fitted parameter is recycled into the prediction, and the modulation amplitude is not defined in terms of the quantity being constrained. The only potentially load-bearing self-citation is Ref. [11], whose authors include M. Giannotti, also an author of this paper. However, that reference is a published, parameter-free theoretical derivation of the transverse-plasmon flux with stated assumptions, and it is independently implemented in the open-source SolarAxionFlux package; it is not a result that is only asserted by the present authors to make the conclusion forced. The solar magnetic-field model is assumed a priori, chosen to saturate helioseismic upper bounds, and the paper explicitly states it 'should not be interpreted as measurements of the internal solar magnetic field.' This makes the quoted limit conditional on that model, and the omission of the overall normalization uncertainty is a legitimate physics concern, but it is not a circularity: the magnetic-field strength is an external input, not an output of the analysis. Thus the central claim—first experimental bounds including the transverse-plasmon flux—is a genuine measurement-based limit under a stated model assumption. No step reduces by construction to its own inputs; any inadequacy in the magnetic-field model is a systematic uncertainty rather than a circular derivation.
Axiom & Free-Parameter Ledger
free parameters (2)
- Peak magnetic-field strengths in the seismic solar model =
B_RZ ≈ 3e7 G, B_tachocline ≈ 5e5 G, B_convective ≈ 5e4 G
- Per-detector background nuisance parameters R0,d and f_d =
Not reported individually
axioms (5)
- domain assumption Resonant transverse-plasmon production formula Eq. (3), taken from refs. [11,19], correctly describes the solar axion flux including finite-width effects and weak mixing.
- domain assumption The solar magnetic-field model saturating helioseismic upper bounds is representative, with ±50% uncertainty.
- domain assumption The inverse Primakoff cross section, Eq. (6), in the relativistic limit applies to Na and I nuclei in the 1–20 keV region of interest.
- domain assumption The time-dependent signal template Eq. (8) with normalization Eq. (9) correctly models the annual modulation from the Earth–Sun distance variation.
- domain assumption The Geant4-based background template plus a flat residual term describes the detector time evolution, with the sub-3 keV unexplained component being time-independent.
read the original abstract
Solar axion-like particles (ALPs) can be resonantly produced through the conversion of transverse plasmons in the magnetic field of the solar interior, introducing a production mechanism complementary to the conventional Primakoff process. In this Letter, we report the first experimental search for solar ALPs produced through this mechanism. Using an exposure of 625.75 kg$\times$yr from ANAIS-112, we search for the annual modulation of the flux induced by the variation of the Earth--Sun distance, assuming detection through inverse Primakoff conversion in the detector. Modulation amplitudes fitted in 1 keV bins between 1--20 keV are consistent with the absence of modulation, yielding the first constraints on the axion--photon coupling that include the resonant transverse-plasmon flux, $g_{a\gamma} < 1.32\times 10^{-9} \mathrm{GeV}^{-1}$ (90\% C.L.) at $m_a \simeq 150$ eV. This previously unexplored production channel extends the sensitivity of ANAIS-112 in the mass range 35--280 eV, where, for the expected solar magnetic-field models, the transverse-plasmon flux exceeds the Primakoff flux by up to two orders of magnitude.
Figures
Reference graph
Works this paper leans on
-
[1]
R. D. Peccei and H. R. Quinn, Phys. Rev. Lett.38, 1440 (1977)
1977
-
[2]
Weinberg, Phys
S. Weinberg, Phys. Rev. Lett.40, 223 (1978)
1978
-
[3]
Wilczek, Phys
F. Wilczek, Phys. Rev. Lett.40, 279 (1978)
1978
-
[4]
J. Jaeckel and A. Ringwald, Annual Review of Nuclear and Particle Science60, 405 (2010), arXiv:1002.0329 [hep-ph]
Pith/arXiv arXiv 2010
-
[5]
D. J. E. Marsh, Phys. Rept.643, 1 (2016), arXiv:1510.07633 [astro-ph.CO]
Pith/arXiv arXiv 2016
-
[6]
I. G. Irastorza and J. Redondo, Prog. Part. Nucl. Phys. 102, 89 (2018), arXiv:1801.08127 [hep-ph]
Pith/arXiv arXiv 2018
-
[7]
Sikivie, Phys
P. Sikivie, Phys. Rev. Lett.51, 1415 (1983)
1983
-
[8]
G. G. Raffelt, Phys. Rev. D33, 897 (1986)
1986
-
[9]
G. G. Raffelt, Phys. Rept.198, 1 (1990)
1990
-
[10]
Redondo, JCAP1312, 008
J. Redondo, JCAP1312, 008
-
[11]
Guarini, M
E. Guarini, M. Gianotti, A. Mirizzi, and J. Redondo, Phys. Rev. D102, 123024 (2020)
2020
-
[12]
G. A. et al. (COSINE-100 Collaboration), Physical Re- view D107, 122004 (2023), arXiv:2302.10267 [hep-ex]
Pith/arXiv arXiv 2023
-
[13]
Anastassopouloset al.(CAST), Nature Physics13, 584 (2017)
V. Anastassopouloset al.(CAST), Nature Physics13, 584 (2017)
2017
-
[14]
CAST Collaboration, Phys. Rev. Lett.133, 221005 (2024)
2024
-
[15]
I. J. Arnquistet al.(Majorana), Physical Review Letters 129, 081803 (2022), arXiv:2206.05789 [hep-ex]
Pith/arXiv arXiv 2022
-
[16]
Zenget al.(PandaX), Physical Review Letters134, 10.1103/physrevlett.134.041001 (2025)
X. Zenget al.(PandaX), Physical Review Letters134, 10.1103/physrevlett.134.041001 (2025)
-
[17]
Aprileet al.(XENONnT), Physical Review Letters 129, 10.1103/physrevlett.129.161805 (2022)
E. Aprileet al.(XENONnT), Physical Review Letters 129, 10.1103/physrevlett.129.161805 (2022)
-
[18]
D. S. Akeribet al., Search for new physics via low-energy electron recoils with a 4.2 tonne×year exposure from the lz experiment (2025), arXiv:2511.17350 [hep-ex]
arXiv 2025
-
[19]
S. Hoof, J. Jaeckel, and L. J. Thormaehlen, Journal of Cosmology and Astroparticle Physics2021(09), 006, arXiv:2101.08789 [hep-ph]
-
[20]
J. N. Bahcall, A. M. Serenelli, and S. Basu, The Astro- physical Journal Supplement Series165, 400 (2006)
2006
- [21]
-
[22]
J. Amar´ eet al., Phys. Rev. Lett.135, 051001 (2025), arXiv:2502.01542 [astro-ph.IM]
arXiv 2025
-
[23]
J. Amar´ eet al., Eur. Phys. J. C79, 228 (2019), arXiv:1812.01472 [astro-ph.IM]
Pith/arXiv arXiv 2019
-
[24]
J. B. Dent, B. Dutta, J. L. Newstead, E. Segreto, and C. Shimmin, Phys. Rev. Lett.125, 131804 (2020), arXiv:2006.01865 [hep-ph]
Pith/arXiv arXiv 2020
-
[25]
K. Abeet al.(XMASS), Phys. Rev. Lett.126, 121801 (2021), arXiv:2011.05267 [hep-ex]
Pith/arXiv arXiv 2021
-
[26]
J. Amareet al.(ANAIS), Eur. Phys. J. C79, 412 (2019), arXiv:1901.09353 [astro-ph.IM]
Pith/arXiv arXiv 2019
-
[27]
J. Amar´ eet al., J. Phys. Conf. Ser.2156, 012175 (2021), arXiv:2110.08109 [physics.ins-det]
Pith/arXiv arXiv 2021
-
[28]
J. Amareet al., Phys. Rev. D103, 102005 (2021), arXiv:2103.01175 [astro-ph.IM]
Pith/arXiv arXiv 2021
-
[29]
The TP-axion flux data has been made available upon request at, https://github.com/RamaAlkaddah/TP Axion analysis ANAIS
-
[30]
Origins Excellence Cluster, Dark Matter Data Cen- ter (2026), https://www.origins-cluster.de/odsl/dark- matter-data-center
2026
-
[31]
H. M. Antia, S. M. Chitre, and M. J. Thompson, Astron. Astrophys.360, 335 (2000), arXiv:astro-ph/0005587
Pith/arXiv arXiv 2000
-
[32]
A. Friedland and A. Gruzinov, Astrophys. J.601, 570 (2004), arXiv:astro-ph/0211377
Pith/arXiv arXiv 2004
-
[33]
S. Couvidat, S. Turck-Chieze, and A. G. Kosovichev, As- trophys. J.599, 1434 (2003), arXiv:astro-ph/0203107
Pith/arXiv arXiv 2003
-
[34]
Redondo and G
J. Redondo and G. Raffelt, Journal of Cosmology and Astroparticle Physics2013(08), 034
-
[35]
Adhikariet al.(COSINE-100), Sci
G. Adhikariet al.(COSINE-100), Sci. Rep.13, 4676 (2023), arXiv:2208.05158 [hep-ex]
Pith/arXiv arXiv 2023
-
[36]
T. Pardo,Advances in detector response and background modeling in the ANAIS-112 experiment for annual mod- ulation and other rare event searches, Ph.D. thesis, U. Zaragoza (2025), arXiv:2510.19469 [astro-ph.IM]
arXiv 2025
-
[37]
Coarasaet al., JCAP11, 048, [Erratum: JCAP 06, E01 (2023)], arXiv:2209.14113 [astro-ph.IM]
I. Coarasaet al., JCAP11, 048, [Erratum: JCAP 06, E01 (2023)], arXiv:2209.14113 [astro-ph.IM]
Pith/arXiv arXiv 2023
-
[38]
I. Coarasaet al., Commun. Phys.7, 345 (2024), arXiv:2404.17348 [astro-ph.IM]. S1 Supplemental Material This Supplemental Material provides the theoretical details supporting the analysis presented in the main Letter. We summarize the solar magnetic-field models adopted throughout this work, review the plasma dis- persion relations governing plasmon–axion ...
Pith/arXiv arXiv 2024
-
[2026]
Color meets Flavor
The present analysis uses the 6.4-year data release, corresponding to an exposure of 625.75 kg×yr [22]. The detectors are surrounded by passive and active shielding, including archaeological and low-activity lead, an anti-radon enclosure, active muon vetoes, polyethy- lene shielding, and water tanks [23]. Detector response has been continuously monitored ...
2010
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.